Several recently published articles report on the utilization of ToxCast data for chemical assessment and prioritization for further testing. Many chemicals highlighted as being of potential concern are currently permitted for use in food contact materials. ToxCast is a large screening program run by the U.S. Environmental Protection Agency (EPA). Within ToxCast, thousands of commonly used chemicals have been tested in hundreds of high-throughput toxicity assays (FPF reported).

In an article published on April 19, 2016 in the peer-reviewed journal Food and Chemical Toxicology,  Agnes Karmaus and colleagues from the U.S. EPA report on food-relevant chemicals that have been tested in ToxCast. First, the authors surveyed various U.S databases to compile a list of 8659 food-relevant chemicals. Of note, metals, polymers and unstable isomers were not included in this list. Subsequent search among 3784 ToxCast chemicals identified 1530 food-relevant substances that have been tested in this program. These comprised 616 direct food additives (DAs), 371 food contact substances (FCSs), and 543 pesticides. Chemicals belonging to more than one of these three categories were counted only once, with priority given to DAs, followed by FCSs, and pesticides.

For most of the 1530 food-relevant chemicals found in ToxCast, data from 300-800 endpoints was available, with minimum and maximum of 95 and 1057 endpoints per chemical, respectively. To distinguish between overt cytotoxicity and selective bioactivity, a ‘cytotoxicity center’ was calculated for each chemical, and bioactivity was flagged as such only if it occurred at concentrations below a chemical-specific cytotoxicity limit. Cytotoxicity center below 1000 µM (the highest concentration tested) was estimated for 10% of DAs, 29% of FCSs, and 40% of pesticides. Four FCSs with cytotoxicity center below 2 µM were:

tributyltin chloride (0.78 µM; CAS 1461-22-9),

dibutyltin chloride (1.07 µM; CAS 683-18-1),

D&C Violet 2 (1.75 µM; CAS 81-48-1),

bis(trichloromethyl)sulfone (1.94 µM; CAS 3064-70-8).

The range of cytotoxic concentrations for FCSs and pesticides was comparable, while cytotoxicity of DAs was somewhat lower. Selective bioactivity was found for 99 FCSs, averaging 10 active assay endpoints per chemical. For comparison, for the two other categories, DAs and pesticides, it was 52 chemicals with 7 endpoints per chemical, and 212 chemicals with 14 endpoints per chemical, respectively.

Another approach utilizing ToxCast data was taken by Scott Auerbach and colleagues from U.S. EPA, collaborating with metabolic disruption experts from several U.S. universities. In their article published on March 15, 2016 in the peer-reviewed journal Environmental Health Perspectives, the researchers highlight a broad range of environmental chemicals that may play a role in obesity and diabetes (FPF reported). Relevant ToxCast assays reflecting several pathways involved in metabolic disruption were identified with the help of experts knowledgeable in the mechanisms of obesity and diabetes. The pathways covered included adipocyte differentiation, feeding behavior in rodents, feeding behavior in nematode, insulin sensitivity in peripheral tissue, pancreatic islet cell function, and pancreatic beta cell function.

The highest activity in these pathways, after filtering for cytotoxicity, was found for pharmaceuticals and pesticides. However, several FCSs were also identified among the top thirty chemicals that were active in one or several pathways (Table 1). Some of these compounds (silicon dioxide, ricinoleic acid, sodium dodecyl sulfate, piperazine) are also registered as DAs. Of interest, most phthalates used in food packaging, as well as bisphenol A (CAS 80-05-7), are also thought to be associated with metabolic disruption (FPF reported). These chemicals were also active in several pathways analyzed by Auerbach et al., but with a lower score, hence their absence from the 30 most active chemicals highlighted for each pathway. Of note, the ability of ToxCast assays to reliably identify obesogenic chemicals has been questioned earlier this year (FPF reported).

Table 1

Chemical nameCAS number
Diallyl phthalate131-17-9
Sodium abietate14351-66-7
Silicon dioxide7631-86-9
Ricinoleic acid141-22-0
Calcium neodecanoate27253-33-4
Glyceryl monoricinoleate1323-38-2
Sodium dodecyl sulfate151-21-3
Sodium 2,4,7-Tri(propan-2-yl)naphthalene-1-Sulfonate1323-19-9

Several more FCSs were found among the top 10 non-pharmaceutical compounds whose activity profile was most similar to that of the substances used as ‘positive’ controls (‘signpost chemicals’) by Auerbach and colleagues (Table 2). Two of these compounds (benzoic acid and 2-phenoxyethanol) are also registered as DAs.

Table 2

‘'signpost' chemical’ FCSs among the top 10 most similar non-pharmaceuticals 
Chemical nameCAS numberChemical nameCAS number
Haloperidol52-86-8Didecyldimethylammonium chloride7173-51-5
Tributyltin chloride1461-22-9
Benzoic acid65-85-0
Diacetone alcohol123-42-2
Tributyltin chloride1461-22-9Didecyldimethylammonium chloride7173-51-5

Finally, a study by Windy Boyd and colleagues from Biomolecular Screening Branch, Division of the National Toxicology Program, National Institute of Environmental Health Sciences (NIEHS), National Institutes of Health (NIH), Department of Health and Human Services (DHHS), North Carolina, U.S., published on May 1, 2016 in the peer-reviewed journal Environmental Health Perspectives, evaluated the effects of 968 ToxCast chemicals on the growth and development of a nematode, Caenorhabditis elegans. Three FCSs, including tributyltin chloride, diphenylamine (CAS 122-39-4), and pyrene (CAS 129-00-0, a contaminant of mineral oil), were found among those having a strong impact on nematode development.

It has to be noted that the activity of a certain chemical in a particular ToxCast assay is not necessarily predictive of a corresponding effect in vivo, hence the information collected by the reviewed studies should be viewed as preliminary and requiring further confirmatory research. These data should primarily aid in hypothesis generation and prioritization of chemicals for follow-up testing.


Karmaus, A., et al. (2016). “Evaluation of food-relevant chemicals in the ToxCast high-throughput screening program.Food and Chemical Toxicology 92:188-196.

Auerbach, S., et al. (2016). “Prioritizing environmental chemicals for obesity and diabetes outcomes research: A screening approach using ToxCast high throughput data.Environmental Health Perspectives (published online March 15, 2016).

Boyd, W., et al. (2016). “Developmental effects of the ToxCastTM Phase I and Phase II chemicals in Caenorhabditis elegans and corresponding responses in zebrafish, rats, and rabbits.Environmental Health Perspectives 124: 586-593.